|
[Sponsors] |
November 5, 2015, 02:27 |
How to define spark plug for combustion
|
#1 |
Member
Pradeep
Join Date: Nov 2014
Posts: 45
Rep Power: 11 |
Do i have to model spark plug or any other options to initiate spark in closed cycle analysis of SI combustion?
|
|
November 9, 2015, 03:16 |
|
#2 |
Senior Member
Blanco
Join Date: Mar 2009
Location: Torino, Italy
Posts: 193
Rep Power: 17 |
Yes for sure, you need an ignition model to initialize spark.
|
|
November 11, 2015, 10:13 |
|
#3 |
Member
Pradeep
Join Date: Nov 2014
Posts: 45
Rep Power: 11 |
how to do that?? could you help me..
|
|
November 11, 2015, 10:20 |
|
#4 |
Senior Member
Blanco
Join Date: Mar 2009
Location: Torino, Italy
Posts: 193
Rep Power: 17 |
It really depends on what combustion model you are using, there isn't a always-valid ignition model...
What combustion model are you going to use? |
|
November 12, 2015, 00:40 |
|
#5 |
Member
Pradeep
Join Date: Nov 2014
Posts: 45
Rep Power: 11 |
ECFM-3Z model.
|
|
November 12, 2015, 03:01 |
|
#6 |
Senior Member
Blanco
Join Date: Mar 2009
Location: Torino, Italy
Posts: 193
Rep Power: 17 |
If you are using ECFM-3Z than you have a built-in ignition model available, look in the IC panel to setup the ignition... You have to specify where the ignition happens and when.
Otherwise you can use AKTIM if you prefer, but it's more complicated. |
|
September 2, 2016, 06:12 |
|
#7 |
New Member
Join Date: May 2016
Posts: 17
Rep Power: 9 |
hello guys
can somone of you explain to me what I am doing wrong? I have an easy simulation without any boundaries. Its only a chamber with 5%-CH4(lambda=1) 21%-O2 and the rest is N2. The temperature is 500K and the pressure is 8bar. Also I am using the ECFM-3Z model, spark. I tried all three ignition models(Standard, AKTIM,ISSIM), but I never get a ignition? What is the problem here? Any suggestions are welcome |
|
September 2, 2016, 07:59 |
|
#8 |
Member
Pradeep
Join Date: Nov 2014
Posts: 45
Rep Power: 11 |
This is the same problem am facing with... Kindly help us...
|
|
September 2, 2016, 09:13 |
|
#9 |
Senior Member
Blanco
Join Date: Mar 2009
Location: Torino, Italy
Posts: 193
Rep Power: 17 |
Guys, it's very difficult to help you on this topic.
The list of possible problems is very long... To start: - double check your lambda distribution around the ignition point at spark timing - double check your ignition point is within the computational domain - double check your spark energy (try to increase it if necessary) Check the .info and .run file for problems. Do you have any "Exctinction" messages? Moreover, what are the concentration you cited? Mass, volume, molar, etc? Because if they're mass concentration then you are not modelling lambda=1 and you are mismatching O2 mass fraction. But if they are volume concentration, then something is wrong in any case: you should have 9.523 mol of air reactiong with 1 mol of CH4 for a stoich. reaction, therefore you should have 1/(1+9.523)=0.095 mol fraction (which is equal to volume fraction for ideal gases). I've done this rapidly, so double check it...and forgive me in case something is wrong! :-) |
|
September 2, 2016, 09:29 |
|
#10 |
New Member
Join Date: May 2016
Posts: 17
Rep Power: 9 |
My concentrations are mass fractions.
Why do you think that 5%(mass.)-CH4 arent lamba=1? And what do you mean that I missmatch the O2 mass fraction? (air consists of ca 23.3% O2 and 76.7% N2 , [mass fraction]) In the literature lamda=1 for methane is 9.5(vol.)%, which is about 5(mass.)%. |
|
September 2, 2016, 09:34 |
|
#11 |
Senior Member
Blanco
Join Date: Mar 2009
Location: Torino, Italy
Posts: 193
Rep Power: 17 |
X_CH4=0.05 g/g gives lambda=1.1, you should have 0.055 mass fraction for lambda =1. Try compute it yourself.
Considering O2, if 5% is CH4 and 21% is O2, then N2 = 74% g/g. Therefore O2_mass_fraction_in_air = 21/(21+74)=0.22 g/g but you should have 23.3% g/g as you know. |
|
September 2, 2016, 09:47 |
|
#12 |
New Member
Join Date: May 2016
Posts: 17
Rep Power: 9 |
Yes you are right. But it shouldnt matter that much because it is still in the range where it should burn.
But in my simulation I dont have even an ignition. So I dont think, that the problem is in the lambda. Also I treid higher spark energy (0.1J for ISSIM) and there was still no ignition. Also I set my ignition point at 0/0/0, which is in the middle of the chamber. (this coordinate coresponds with the initial coordinate system so it has to be in the computational domain, doesnt it?) Maybe something what is mentionable: I only use ProStar and no es-ice. But I dont thnik, that this should be a problem? |
|
September 2, 2016, 09:59 |
|
#13 |
Senior Member
Blanco
Join Date: Mar 2009
Location: Torino, Italy
Posts: 193
Rep Power: 17 |
Yes I know that lambda 1.1 is still in the "flammable" range, but I just wanted to be sure that you setup fits with your expectations.
In any case, it is not trivial that the poit (0,0,0) in csys 0 is inside the computational domain, it depends where your domain is! If (0,0,0) is in the middle of your chamber, than it is ok. I've always used es-ice in my past computations with ecfm-3z, therefore I really don't know if pro-star is enough without any additional user intervention. I mean, as an example, does pro-star creates automatically all the scalars (tracking scalars, passive scalars, active scalars) that are needed by the ecfm-3z model? Moreover I'm pretty sure ecfm-3z setup withit es-ice activates some option that are not easily available in pro-star gui... |
|
September 2, 2016, 10:07 |
|
#14 |
New Member
Join Date: May 2016
Posts: 17
Rep Power: 9 |
The initial coordinate system is in the middle, so I think if I choose the (0/0/0) point it should be fine.
I dont use es-ice, because I only have a pre-chamber where I want to look at the combustion. ProStar generates all the scalars, but I dont have the options like I would have had in es-ice. Nevertheless Im pretty sure, that it should work witout es-ice too |
|
September 2, 2016, 10:33 |
|
#15 |
Senior Member
Blanco
Join Date: Mar 2009
Location: Torino, Italy
Posts: 193
Rep Power: 17 |
Sorry I can't help you further, as far as I remember when I looked at the pro-STAR GUI for the chemical reaction scheme, most of the options were "disabled" (grayed-out) by default, and that was because I preliminary did the ecfm-3z setup in es-ice. Moreover, in pro-STAR there was not a ECFM-3Z model, but instead it was showing something like CFM chemical reaction scheme, even if es-ice was effectively using the ecfm-3z at the end...therefore some not-trivial relation do exists in the relation between es-ice and pro-STAR.
The ignition scheme was also grayed-out, only some numerical parameter were available (but I've always done the setup in es-ice as I mentioned). |
|
September 9, 2016, 03:41 |
Closed cycle analysis
|
#16 |
Member
Pradeep
Join Date: Nov 2014
Posts: 45
Rep Power: 11 |
I have done simulation for closed cycle SI engine .... from 575CAD to 720CAD.... it just contains piston, dome and liner as the file attached... if i need to simulate combustion with same geometry is it possible??? since its a closed simulation, i hope its definitely possible. Could you please define the steps to proceed in STAR controls...(refer file attached)
|
|
September 9, 2016, 08:00 |
|
#17 | |
New Member
Can
Join Date: Jan 2013
Location: Istanbul
Posts: 26
Rep Power: 13 |
Yes, it is possible. This is the way how I run the combustion simulations for Diesel and Otto engines.
I think it could be easier for you to follow documents of Star-Cd for es-ice module which are the user guide and tutorials guide. You can find them in the installation folders. Quote:
|
||
September 12, 2016, 05:01 |
spark plug-dome same region
|
#18 |
Member
Pradeep
Join Date: Nov 2014
Posts: 45
Rep Power: 11 |
i modelled the spark plug at 0,0,0 where the dome region is present..when i check regions at star cpntrols. it shows error saying "1 region present in combustion dome" ...for your referencei have attached the geometry file. ...
How to draw a spark plug without connecting to dome region as in my case?? or else wats the solution for this?? Last edited by pradeep6890; September 30, 2016 at 17:47. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
HELP----Surface Reaction UDF | Ashi | Fluent UDF and Scheme Programming | 1 | May 19, 2020 21:13 |
SR_Rate UDF For a MEMs model | timjm | Fluent UDF and Scheme Programming | 1 | October 10, 2018 02:16 |
udf problem | eb.nabizadeh | Fluent UDF and Scheme Programming | 2 | March 1, 2013 00:28 |
Installing OF 1.6 on Mac OS X | gschaider | OpenFOAM Installation | 129 | June 19, 2010 09:23 |
Free surface boudary conditions with SOLA-VOF | Fan | Main CFD Forum | 10 | September 9, 2006 12:24 |